Google’s YouTube platform is rolling out major changes to how it identifies and displays AI-generated content, introducing automatic detection capabilities and more visible labels for viewers.
Variety reports that YouTube has announced significant updates to its approach to labeling AI-generated content, introducing automatic detection systems and making disclosures more prominent for viewers. The changes represent an evolution of policies the platform first implemented in 2024, when it began requiring creators to manually disclose their use of AI tools.
The video giant revealed the updates in a blog post, stating that the modifications are designed to provide greater transparency while maintaining creator control. Starting this week, YouTube will deploy a new internal system capable of automatically identifying videos that make significant use of photorealistic AI technology.
Under the new system, creators remain required to manually disclose when they use realistic AI in their content. However, YouTube will now automatically apply labels to videos where its detection systems identify significant photorealistic AI use, even if creators have not made such disclosures themselves. According to the company, this automatic labeling will help ensure viewers can quickly identify content that appears real but was actually generated or significantly altered using artificial intelligence.
Creators who believe their content has been incorrectly flagged as AI-generated will have the option to modify the disclosure status through the YouTube Studio tool. However, YouTube has specified that labels will remain permanent in certain cases. These include content created using YouTube’s own AI tools, such as Veo or Dream Screen, and content containing C2PA metadata from the Coalition for Content Provenance and Authenticity that indicates full AI generation.
Beyond detection capabilities, YouTube is also changing where AI content labels appear to make them more immediately visible to viewers. Previously, the platform placed AI content disclosures in the expanded description section of videos. Under the new system, labels for photorealistic and meaningfully AI-altered content will occupy much more prominent positions.
Breitbart News previously reported that 20 percent of the new videos on YouTube are AI slop, with much of the low quality content aimed at children:
A study conducted by video-editing company Kapwing has revealed that more than 20 percent of the videos recommended to new YouTube users are “AI slop” — low-quality, AI-generated content designed to garner views and monetize attention. The researchers surveyed 15,000 of the most popular YouTube channels worldwide and discovered that 278 of them exclusively feature AI slop.
Collectively, these AI slop channels have accumulated over 63 billion views and 221 million subscribers, generating approximately $117 million in yearly revenue. When the researchers created a new YouTube account, they found that 104 of the first 500 recommended videos were AI slop, while one-third of the content fell under the “brainrot” category, which includes both AI slop and other low-quality, attention-grabbing videos.
Read more at Variety here.
Lucas Nolan is a reporter for Breitbart News covering issues of AI, free speech, and online censorship.